Organelle transport along microtubules in Xenopus melanophores: evidence for cooperation between multiple motors

Abstract

Xenopus melanophores have pigment organelles or melanosomes which, in response to hormones, disperse in the cytoplasm or aggregate in the perinuclear region. Melanosomes are transported by microtubule motors, kinesin-2 and cytoplasmic dynein, and an actin motor, myosin-V. We explored the regulation of melanosome transport along microtubules in vivo by using a new fast-tracking routine, which determines the melanosome position every 10 ms with 2-nm precision. The velocity distribution of melanosomes transported by cytoplasmic dynein or kinesin-2 under conditions of aggregation and dispersion presented several peaks and could not be fit with a single Gaussian function. We postulated that the melanosome velocity depends linearly on the number of active motors. According to this model, one to three dynein molecules transport each melanosome in the minus-end direction. The transport in the plus-end direction is mainly driven by one to two copies of kinesin-2. The number of dyneins transporting a melanosome increases during aggregation, whereas the number of active kinesin-2 stays the same during aggregation and dispersion. Thus, the number of active dynein molecules regulates the net direction of melanosome transport. The model also shows that multiple motors of the same polarity cooperate during the melanosome transport, whereas motors of opposite polarity do not compete.

FIGURE 1

Determination of local velocities. The trajectory and individual coordinates obtained for the melanosome were classified under minus-end run (green), plus-end run (blue), or oscillations in position (red) according to the criteria described (scale bar, 0.5 μm). (A). The fragment corresponding to the minus-end run (black line) was divided into segments of 40 points and a linear equation was fitted to each of these segments (green line). From the slope of the best-fitting equation, the velocities were obtained for each of the segments (red). The bottom part of B shows the residuals obtained from the fitting. The velocity data were binned and included in the corresponding histogram (C).

FIGURE 2

Tracking performance. A nanometric stage placed on top of the microscope was programmed to move a slide containing a dried suspension of microspheres in steps of 10–50 nm. Movies of the beads were registered as described in Materials and Methods at 200 frames/s and the particle trajectories were obtained by using the pattern-recognition algorithm. The trajectory of one of the beads moving in steps of 10 nm is plotted as a function of time. (Inset) Size of the step obtained by analyzing the particles trajectories as a function of the input step.

FIGURE 3

Tracking melanosomes in cells. The distance traveled by one of the melanosomes in a cell stimulated with melatonin was measured as described in the text and is represented as a function of time. The continuous lines represent the fit of a linear equation in two regions of the trajectories. From the fitting, we calculated the following velocities in the segments: 0.922 ± 0.003 (I) and 0.551 ± 0.001 (II) μm/s. (Inset) Initial frame of a movie recorded for the same cell at 100 frames/s. The melanosomes appears in the image as black circles. The trajectory of the chosen melanosome is overlapped with the image (black line). (Scale bar,1 μm.)

FIGURE 4

Distribution of velocities of melanosomes. Movies of different regions of two cells stimulated for aggregation (A) or dispersion (B) were recorded as described and the organelle trajectories and velocities calculated. The histograms of velocities were constructed from data of melanosomes moving toward the minus (A) or plus (B) ends of the microtubules. The continuous lines correspond to the fitting of Eq. 5 with the following best-fitting parameters: v₁ = 0.250 ± 0.009 μm/s, σ = 0.078 ± 0.006 μm, A₁ = 1520 ± 240, A₂ = 1230 ± 220, A₃ = 360 ± 140, and A₄ = 240 ± 140 (A); and v₁ = 0.260 ± 0.007 μm/s, σ = 0.098 ± 0.007 μm/s, A₁ = 000 ± 300, A₂ = 700 ± 140, A₃ = 115 ± 113, and A₄ = 60 ± 100 (B). The gray lines show the contribution of each peak to the total distributions. The inset to A shows the AIC values calculated by fitting Eq. 5 to the histogram considering n values from 1 to 6.

FIGURE 5

Tracking melanosomes moving along a single microtubule. Cells with microtubules labeled with EGFP-XTP were treated with latrunculin B and stimulated for pigment aggregation with melatonin. Two-photon fluorescence images of the cells were taken (excitation wavelength = 915 nm) and the melanosomes, detected as bright spots in these images, were tracked as described above. (A and C) The trajectories recovered for two of the melanosomes moving toward the minus end of the microtubules are superimposed over the fluorescence images taken for the cells. Scale bar, 1 μm. The distance traveled by the melanosomes in different regions of these trajectories (arrows) was plotted as a function of time (B and D, respectively). The continuous lines show the fitting of a linear equation to different segments of the trajectories. The velocities (in nm/s) and r² values were 255 ± 2, 0.9988 (I); 216 ± 3, 0.9936 (II); 438 ± 3, 0.9993 (III); 406 ± 2, 0.9963 (IV); 480 ± 7, 0.9941 (V); and 132 ± 1, 0.9959 (VI).

FIGURE 6

Imaging melanophores. Comparison between brightfield (A) and two-photon excitation fluorescence (B) images of a melanophore revealed that melanosomes are observed under two-photon excitation. The excitation wavelength was 780 nm. (Scale bars, 5 μm).

FIGURE 7

Autocorrelation analysis. A histogram of dynein-driven melanosomes obtained after stimulating the cells for aggregation was finely binned and filtered as described in the text until a unimodal curve (inset) was obtained. The difference function resulting from subtraction of the unimodal curve from the histogram was lightly smoothed, and the autocorrelation of the smoothed difference function was calculated. The arrow points the valley and peak used to calculate the AC score.

FIGURE 8

Relative population of melanosomes attached to a different number of active motors. The relative amplitude of the peaks in the normalized histograms corresponding to melanosomes attached to dynein (A) and kinesin-2 (B) motors is represented as a function of the number of active motors during aggregation (solid symbols) and dispersion ( open symbols).